Objective Evaluation of Pedestrian and Vehicle Tracking on the CLEAR Surveillance Dataset

Authors:
Murtaza Taj;Emilio Maggio;Andrea Cavallaro
Affiliations:
Queen Mary, University of London, London, United Kingdom E1 4NS;Queen Mary, University of London, London, United Kingdom E1 4NS;Queen Mary, University of London, London, United Kingdom E1 4NS
Venue:
Multimodal Technologies for Perception of Humans
Year:
2008

Citing 14
Cited 0

Learning Patterns of Activity Using Real-Time Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiresolution adaptive filtering of signal-dependent noise based on a generalized Laplacian pyramid

ICIP '97 Proceedings of the 1997 International Conference on Image Processing (ICIP '97) 3-Volume Set-Volume 1 - Volume 1
Detecting Pedestrians Using Patterns of Motion and Appearance

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
A Noniterative Greedy Algorithm for Multiframe Point Correspondence

IEEE Transactions on Pattern Analysis and Machine Intelligence
Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1 - Volume 01
Principal Axis-Based Correspondence between Multiple Cameras for People Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
An Experimental Study on Pedestrian Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
Interaction between high-level and low-level image analysis for semantic video object extraction

EURASIP Journal on Applied Signal Processing
The AIT outdoors tracking system for pedestrians and vehicles

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
Evaluation of USC human tracking system for surveillance videos

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
Multi-feature graph-based object tracking

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
Multiple vehicle tracking in surveillance videos

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
Robust appearance modeling for pedestrian and vehicle tracking

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
Robust vehicle blob tracking with split/merge handling

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships

Quantified Score

Hi-index	0.00

Visualization

Abstract

Video object detection and tracking in surveillance scenarios is a difficult task due to several challenges caused by environmental variations, scene dynamics and noise introduced by the CCTV camera itself. In this paper, we analyse the performance of an object detector and tracker based on background subtraction followed by a graph matching procedure for data association. The analysis is performed based on the CLEAR dataset. In particular, we discuss a set of solutions to improve the robustness of the detector in case of various types of natural light changes, sensor noise, missed detection and merged objects. The proposed solutions and various parameter settings are analysed and compared based on 1 hour 21 minutes of CCTV surveillance footage and its associated ground truth and the CLEAR evaluation metrics.